According to Professor László Komlósi and Professor Zoltán Baracskai (Szeedsm Doctoral Program), based on a transdisciplinary approach, the essence of the doctoral program, is to create space for knowledge and a research milieu for problems arising from practice, where complex problems can be solved. The relationship between human and machine is in fact the coexistence of natural and artificial intelligence. One of the many areas, which our doctoral candidates have already worked, is working and will continue to work in the future. Within the program we bring every upcoming issue to be researched into one of our Living-Laboratories. For the similar and more frequent problems we established and operate a research groups with our firm partners, students and alumni. Each problem is supervised by a domestic and a foreign mentor.

According to Gábor Bereczki (UpScale), Upscale Labs serves customers on two continents utilizing state of the art technologies in machine learning. In the competitive landscape where we operate it is imperative to stay at the bleeding edge of scientific research. This is why we started our collaboration program with the Szechenyi István University. Cooperation between PhD programs and companies like Upscale in the field of AI has the potential to lead to significant benefits for both parties. PhD programs can gain practical experience and exposure to the latest industry trends and technologies, while companies can access the latest research and developments in AI and recruit talented PhD students who have a deep understanding of AI and related technologies. Furthermore, such cooperation can provide exposure to real-world data and problems, and opportunities to solve complex problems and improve business operations through the use of AI.

In order for such cooperation to be successful, both parties need to approach the collaboration with openness, transparency, and clear communication. This requires careful consideration of academic requirements, business goals and priorities, resources and support for PhD students, and ethical considerations and data privacy when working with company data.

Overall, cooperation between PhD programs and companies in the field of AI can drive innovation and progress, and lead to breakthroughs and advancements that benefit society. By combining academic expertise with industry resources and real-world data, researchers can develop AI systems that address complex problems and contribute to the advancement of the field of AI.

Student reports based on systematic points are available here:

József Pap:

using public data to detect fraud payments

Literature Review and gap analysis, comparing machine learning and. expert system, pilot

Activities so far:
consultations and research activities: one personal and three online consultations. The literature review of the topic (51 articles has been reviewed) is based on the phrases provided by Upscale, narrowing the literature (36 in the period of 2020-2023). Upscale’s request was to find some public date sources as well, so I have forwarded them about 2-3 types of databases from the literature.

Further actions:
Currently I am waiting for the feedback and review of UpScale. We have also agreed on two other topics: Log analysis, and what we have proposed originally in addition to fraud, credit and deposit assessment.

Effects of this collaboration on my research activity:

This is another area where I am able to collaborate in the practical use of AI. With data scientists and mathematician colleagues we might be able to work out an article.
Comments: the whole progress was hindered a bit because of the change of the contact person, but the collaboration and topic are both interesting.

Judit Bilinovics-Sipos: 

Log analysis – error detection – recognizing stagnant processes 

Literature review
Comparative analysis
Interview with an expert
Tentative problem solving process

Activities so far:
all kinds of cosultation and research activities
1: Examining the solutions used for log analysis ont he market, researching the practical use by researching the literature and involving experts
Dynatrace – AI based log aggregator
Splunk – Log aggregator
Elastic – Log aggregator
CloudFabrix – Log aggregator
2: Personal consultation about 4 log aggregator operation and further research directions
The Log aggregator used by Upscale had been presented in action.
Phone consultation about further actions 

Further actions:
Over the last phone consultation, it was suggested to conduct an interview with the relevant developers who are involved in the error detections, in order to understand how they are detected and trakced. 

Effect on my research activity:
During the collaboration, both the literature and the relevant problem area are contributing to mjy research activity, as I examine the mental model of decision makers during human-machine relationships. This Log analysis, error detection and the meaning or non-meaning attributed to the appearing error are also part of my own research. In my publications, despite the fact that it doesn’t deal with this exact research question, the literature can be used. Understanding the phenomenon of the hybrid reality contributes to narrow down the problem area of my research.

Usman Ghani: 

Topic:  Public Data Integration for Financial Open Data in Canadian Market 

Methodology:  Public Data Inventory 

Activities happened:
Shared Folders, Email Communication, Shared case studies, Data Research and Reports, and Submitted reports.
Data Inventor Sheet, Indexes on SDGs, Canadian Research Studies from the indicated sources of Interest, Research Reference Report, WIP Draft

 Further plans:
Alignment on Methodology sophistication and value addition and touch up the current draft. Writing up the Data Analysis, Results and Conclusion.

 Effects on my research activity:
Identification of Possibilities in public Data Research.
Broad introduction with Research Possibilities.
Usability of public data for companies in market analysis.
A conference paper in AS-IS condition can be published.
Novel research and methodological output are expected.

Other notes:
Segregation of proprietary and public data of use for customer and market analysis
Data Sources are narrowed down, and we are close to finalising the data for final analysis and write-up.


Share This